Circadian Content Analysis

This notebook fosters the reproducibility of the following paper:

• Coelho J, Martin VP, Gauld C, D'Incau E, Geoffroy PA, Bourgin P, Philip P, Taillard J, Micoulaud-Franchi JA, The clinical physiology of circadian rhythm: a systematic and hierarchized content analysis of the circadian questionnaires. Under review. 2024.

If you want to compute the same metrics with you own data, we redirect you towards our generic content analysis GitHub https://github.com/vincentpmartin/generic_content_analysis

⚠️ PLEASE READ BEFORE DOING ANYTHING ⚠️

Welcome to this online coding environment ! You are currently running a Jupyter notebook that we hope to be usefull for content analysis of questionnaires.

Each cell can be ran individually (pressing the "Play" ▶ button or pressing CTRL+ENTER in the each cell). In order to make everything run smoothly, it is strongly recommended to run them in order.

RECOMMANDED: You can also run all the cells at the same time : "Run" → "Run All Cells".

At the end of the execution, you will have the opportunity to save all yours results in an HTML file, executing "File" → "Save and Export Notebook as" → "HTML"

If you have chosen to run every cell individually, please begin by running the following cell that imports all the required package for our code to work correctly ↓

	Category	Subcategory	Ab	Symptom	MEQ	MCTQ	CSM	rMEQ	CQ	DTQ	CTQ	CIRENS	CCQ	CTI	CAPS TAQ	SCRAM	SIC	SRQ
0	Circadian phase	Evening time	S01	Bedtime (phase)	1	1	1	0	0	1	0	0	0	0	1	0	0	0
1	Circadian phase	Morning time	S02	Rise time (phase)	1	0	1	0	2	1	1	0	2	0	1	0	0	0
2	Circadian phase	Evening time	S03	Sleep time (phase)	2	1	2	2	0	2	1	0	0	0	0	2	0	0
3	Circadian phase	Morning time	S04	Awake time (phase)	0	1	2	0	0	0	0	0	0	0	0	1	0	0
4	Circadian phase	Daytime	S05	Social activity (phase)	0	0	0	0	0	0	0	0	0	0	1	0	0	0

	Category	Subcategory	Ab	Symptom	CAPS TAQ	CTQ	CQ	MEQ	CSM	MCTQ	CCQ	DTQ	CTI	SCRAM	rMEQ	SRQ	SIC	CIRENS	sum_symptoms
8	Circadian phase	Daytime	S09	Peak feeling (phase)	1	1	1	1	2	0	1	0	0	0	1	0	0	1	8
1	Circadian phase	Morning time	S02	Rise time (phase)	1	1	2	1	1	0	2	1	0	0	0	0	0	0	7
2	Circadian phase	Evening time	S03	Sleep time (phase)	0	1	0	2	2	1	0	2	0	2	2	0	0	0	7
39	Circadian complaints	Evening time	S40	Evening sleepiness (complaint)	0	0	2	2	2	0	0	2	2	2	2	0	0	0	7
11	Circadian phase	Self-evaluation	S12	Self-evaluation (phase)	0	0	0	1	1	0	0	1	0	1	1	0	2	0	6

---

1. Analysis of the number and frequency of symptoms

In a first step, we analyse the frequency of the symptoms.

Histogram of number of symptoms

Sorted by number of occurences

The figure has been save in the online folder (📁 symbol on the left) under the name figure1_histogram_1.pdf.
You can change the name and the format of the file changing the name in the #fig.write_image() function.
⚠️ If you need it, save the figure on your local computer : these online file will be deleted as soon as you quit this page!

Sorted by category

The figure has been save in the online folder (📁 symbol on the left) under the name figure2_histogram_2.pdf.
You can change the name and the format of the file changing the name in the #fig.write_image() function.
⚠️ If you need it, save the figure on your local computer : these online file will be deleted as soon as you quit this page!

Number of symptoms by questionnaire

	Specific symptoms	Compound symptoms	Total
CAPS TAQ	12	4	16
CTQ	9	5	14
CQ	8	5	13
MEQ	9	3	12
CSM	6	6	12
MCTQ	11	0	11
CCQ	9	1	10
DTQ	4	4	8
CTI	3	5	8
SCRAM	4	2	6
rMEQ	3	2	5
SRQ	4	0	4
SIC	0	3	3
CIRENS	1	0	1

The table has been save in the online folder (📁 symbol on the left) under the name table1_symptoms_per_questionnaire.xlsx.
You can change the name and the format of the file changing the name in the sympt_per_questionnaire.to_excel() function.
⚠️ If you need it, save the excel file on your local computer : these online file will be deleted as soon as you quit this page!

Number of symptoms in each category for each questionnaire

	CAPS TAQ	CTQ	CQ	MEQ	CSM	MCTQ	CCQ	DTQ	CTI	SCRAM	rMEQ	SRQ	SIC	CIRENS
Circadian phase	8	3	5	9	8	3	5	4	0	4	3	0	1	1
Circadian amplitude&stability	8	7	5	1	0	0	4	0	4	0	0	0	1	0
Circadian timing	0	0	0	0	0	5	0	0	0	0	0	0	0	0
Circadian regularity	0	0	0	0	0	0	0	0	0	1	0	4	0	0
Circadian complaints	0	4	3	2	4	3	1	4	4	1	2	0	1	0

The table has been save in the online folder (📁 symbol on the left) under the name table2_categorie_per_questionnaire.xlsx.
You can change the name and the format of the file changing the name in the cat_per_questionnaire.T.to_excel() function.
⚠️ If you need it, save the excel file on your local computer : these online file will be deleted as soon as you quit this page!

Distribution across the categories of the symptoms measured by each questionnaire

(i.e. same thing as before, but normalized by questionnaire (sum across lines equals 1)).

Figure 3 has been save in the online folder (📁 symbol on the left) under the name figure3_heatmap.pdf. You can change the name and the format of the file changing the name in the #fig.write_image() function.
⚠️ If you need it, save the figure on your local computer : these online file will be deleted as soon as you quit this page!

---

2. Analysis and data vizualisation of content analysis Figure

Content Analysis Figure

The figure has been save in the online folder (📁 symbol on the left) under the name figure4_radial.pdf. You can change the name and the format of the file changing the name in the #fig.write_image() function.
⚠️ If you need it, save the figure on your local computer : these online file will be deleted as soon as you quit this page!

Overlap between questionnaires - Jaccard Index

In order to estimate the overlap between the symptoms measured by the questionnes, calculate the Jaccard index, which is defined as the number of symmtoms that are measured by both questionnaires, divided by the number of unique symptoms measured both questionnaires.

Jaccard index of symptom for each pair of questionnaire

First, we compute the Jaccard index for each pair of questionnaires and plot it using a heatmap.

	CAPS TAQ	CTQ	CQ	MEQ	CSM	MCTQ	CCQ	DTQ	CTI	SCRAM	rMEQ	SRQ	SIC	CIRENS
CAPS TAQ	1.000000	0.304348	0.318182	0.333333	0.217391	0.038462	0.368421	0.090909	0.142857	0.047619	0.050000	0.000000	0.000000	0.062500
CTQ	0.304348	1.000000	0.350000	0.181818	0.181818	0.136364	0.200000	0.222222	0.294118	0.052632	0.117647	0.000000	0.062500	0.071429
CQ	0.318182	0.350000	1.000000	0.388889	0.315789	0.000000	0.642857	0.166667	0.312500	0.117647	0.200000	0.000000	0.000000	0.076923
MEQ	0.333333	0.181818	0.388889	1.000000	0.600000	0.095238	0.294118	0.333333	0.176471	0.285714	0.416667	0.000000	0.071429	0.083333
CSM	0.217391	0.181818	0.315789	0.600000	1.000000	0.210526	0.222222	0.538462	0.176471	0.285714	0.416667	0.000000	0.071429	0.083333
MCTQ	0.038462	0.136364	0.000000	0.095238	0.210526	1.000000	0.000000	0.266667	0.055556	0.133333	0.066667	0.000000	0.000000	0.000000
CCQ	0.368421	0.200000	0.642857	0.294118	0.222222	0.000000	1.000000	0.058824	0.200000	0.066667	0.153846	0.000000	0.000000	0.100000
DTQ	0.090909	0.222222	0.166667	0.333333	0.538462	0.266667	0.058824	1.000000	0.142857	0.272727	0.300000	0.000000	0.100000	0.000000
CTI	0.142857	0.294118	0.312500	0.176471	0.176471	0.055556	0.200000	0.142857	1.000000	0.076923	0.181818	0.000000	0.100000	0.000000
SCRAM	0.047619	0.052632	0.117647	0.285714	0.285714	0.133333	0.066667	0.272727	0.076923	1.000000	0.375000	0.111111	0.125000	0.000000
rMEQ	0.050000	0.117647	0.200000	0.416667	0.416667	0.066667	0.153846	0.300000	0.181818	0.375000	1.000000	0.000000	0.142857	0.200000
SRQ	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.111111	0.000000	1.000000	0.000000	0.000000
SIC	0.000000	0.062500	0.000000	0.071429	0.071429	0.000000	0.000000	0.100000	0.100000	0.125000	0.142857	0.000000	1.000000	0.000000
CIRENS	0.062500	0.071429	0.076923	0.083333	0.083333	0.000000	0.100000	0.000000	0.000000	0.000000	0.200000	0.000000	0.000000	1.000000

Table 3 has been save in the online folder (📁 symbol on the left) under the name table3_jaccard_pairs.xlsx.
You can change the name and the format of the file changing the name in the jaccard_table.to_excel() function.
⚠️ If you need it, save the excel file on your local computer : these online file will be deleted as soon as you quit this page!

The figure has been save in the online folder (📁 symbol on the left) under the name figure5_heatmap_jaccard.pdf.
You can change the name and the format of the file changing the name in the #fig.write_image() function.
⚠️ If you want it, save the figure on your local computer : these online file will be deleted as soon as you quit this page!

Avg. Jaccard index

Then, we compute the average of Jaccard index for each questionnaire with other questionnaires (excluding the references).

	Avg. Jaccard Index
CAPS TAQ	0.151848
CTQ	0.167300
CQ	0.222266
MEQ	0.250796
CSM	0.255371
MCTQ	0.077139
CCQ	0.177458
DTQ	0.191744
CTI	0.143044
SCRAM	0.150007
rMEQ	0.201628
SRQ	0.008547
SIC	0.051786
CIRENS	0.052117

Average Jaccard index (wo references): 0.15007 (sd: 0.0767)

Table 4 has been save in the online folder (📁 symbol on the left) under the name table4_jaccard_average_questionnaires.xlsx.
You can change the name and the format of the file changing the name in the jaccard.to_excel() function.
⚠️ If you need it, save the excel file on your local computer : these online file will be deleted as soon as you quit this page!

Correlation between the number of symptoms and the average Jacquart index for each questionnaire

	Avg. Jaccard Index	Specific symptoms	Compound symptoms	Total
CAPS TAQ	0.151848	12	4	16
CTQ	0.167300	9	5	14
CQ	0.222266	8	5	13
MEQ	0.250796	9	3	12
CSM	0.255371	6	6	12
MCTQ	0.077139	11	0	11
CCQ	0.177458	9	1	10
DTQ	0.191744	4	4	8
CTI	0.143044	3	5	8
SCRAM	0.150007	4	2	6
rMEQ	0.201628	3	2	5
SRQ	0.008547	4	0	4
SIC	0.051786	0	3	3
CIRENS	0.052117	1	0	1

Correlation between Jaccard Index and number of specific symptoms:  SignificanceResult(statistic=0.34851905254334287, pvalue=0.222007239925357)
Correlation between Jaccard Index and number of compound symptoms:  SignificanceResult(statistic=0.5695643285911823, pvalue=0.0334921506582554)
Correlation between Jaccard Index and total number of symptoms:  SignificanceResult(statistic=0.5792965594550208, pvalue=0.029933781366618964)

Jaccard index of symptoms for each pair of questionnaire for each category

Computing the same metric (average of average) for each category of questionnaires.

c:\Users\Vincent\anaconda3\lib\site-packages\numpy\core\fromnumeric.py:3440: RuntimeWarning:

Mean of empty slice.

c:\Users\Vincent\anaconda3\lib\site-packages\numpy\core\_methods.py:189: RuntimeWarning:

invalid value encountered in double_scalars

	Avg. Jaccard Index
Circadian phase	0.254176
Circadian amplitude&stability	0.223167
Circadian timing	NaN
Circadian regularity	0.250000
Circadian complaints	0.257316

Table 5 has been save in the online folder (📁 symbol on the left) under the name table5_jaccard_categories.xlsx.
You can change the name and the format of the file changing the name in the res.to_excel() function.
⚠️ If you need it, save the excel file on your local computer : these online file will be deleted as soon as you quit this page!

Sunburst Plots

Unweighted sunburst plot

The figure has been save in the online folder (📁 symbol on the left) under the name figure6_sunburst_plot.pdf.
You can change the name and the format of the file changing the name in the #fig.write_image() function.
⚠️ If you want it, save the figure on your local computer : these online file will be deleted as soon as you quit this page!

Sunburst weighted by number of occurence of symptoms

The figure has been save in the online folder (📁 symbol on the left) under the name figure7_sunburst_plot.pdf.
You can change the name and the format of the file changing the name in the #fig.write_image() function.
⚠️ If you want it, save the figure on your local computer : these online file will be deleted as soon as you quit this page!

---

Export to html

You have reached the end of this notebook. If you want to save the whole page, you can download it to html with dynamic figures:

• "File" → "Save and Export Notebook as" → "HTML"